On Oct 30, 9:24 pm, [EMAIL PROTECTED] (Jenda Krynicky) wrote:
> On 30 Oct 2007 at 10:22, Paul Lalli wrote:
>
>
>
> > On Oct 30, 11:15 am, [EMAIL PROTECTED] (Chas. Owens) wrote:
> > > On 10/30/07, Kaushal Shriyan <[EMAIL PROTECTED]> wrote:
> > > snip> Whats the exact purpose of use strict
>
> > > snip
>
> > > The strict pragma has three effects (unless modified by arguments):
> > > 1. most variables must be declared (there are some exceptions)
>
> > This is a common misperception.  use strict 'vars' in fact does not
> > require anything to be declared.  All it requires is that global
> > variables must be fully qualified.  You cannot use the short-name of
> > globals that belong to the current package.
>
> > <snipped>
>
> > With strict enabled, the only way to use a short-name of a variable is
> > to declare a lexical of that name (the right choice) using 'my', or to
> > disable strict 'vars' on a variable-by-variable case using 'our' (the
> > wrong choice).  This is what leads people to assert "use strict forces
> > you to declare your variables".
>
> Technicaly you are right,

Actually he's subtly wrong. But that's not important.

> but generally it's better to present use
> strict 'vars' as a way to force oneself to declare variables. It may
> not be exactly what it does, but it's why it exists.

I prefer to think of (and initially describe) 'strict' as switching
off three optional (and only occasionally useful) extensions to the
language that you wouldn't want Perl to accidentally think you were
trying to use when you make a mistake.

As users become more experienced they will reach a point where these
extensions may be useful to them. And at that point, and only at that
point, is it worth trying to understand in detail what 'no strict'
does.


-- 
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
http://learn.perl.org/


Reply via email to